Welcome to the Q&A Forum

Hello,

The source code on those pages have a Robots Index,Follow tag. This should be noindex, follow. They also have two rel = canonical tags, which point to different URLs, thus conflicting with each other.

Really I would have to log into the back-end to see what's going on, but I'm guessing one of the rel canonical tags is being added by Mageworks or Yoast and the other is default from Wordpress. You may require some custom coding to the blog template, and I recommend working with a developer if you can't figure it out on your own. There are lots of Wordpress developers on sites like Upwork and Guru who could probably get this done for you for a minimal investment.

Hello Geraldine,

I'm going to rephrase your questions here just to make sure I understand them correctly.

1. Your Sort-By URLs are using self-referencing rel= canonical tags.

When I looked into this issue I found that some of these pages no longer exist and produce a 404 error page. Example:
https://www.tidy-books.co.uk/childrens-bookcases-shelves/sort-by/position/sort-direction/desc/l/letters:5

I did find some that I was able to access and verified the problem above. You should contact MageWorks to ask if there is a way you can fix this with the SEO plugin you're using. If not, I suggest adding a disallow statement in the robots.txt file for the /sort-by/ folders to keep them from being accessed and indexed. Then I would go into Google Search Console (GSC) and remove that entire directory from the index, as described here: https://support.google.com/webmasters/answer/1663419?hl=en .

2. You have duplicate title tags, mostly caused by the blog and the sort-by URLs discussed above.

If you're not using the Yoast SEO plugin for Wordpress you should be. Without seeing the report and knowing exactly what those URLs are with duplicate title tags, I can't help answer this question very well. Common examples are Tag pages and paginated Category pages. I usually advise blog owners to disallow the Tag pages in the Robots.txt file and removing the directory from Google SERPs using GSC URL Removal Tool, as described above.

3. Image directory pages are causing mobile usability errors in GSC and you have blocked them from being indexed.

Because those index pages do not need to be accessed in order to render the page, I don't see any problem with blocking those in the robots.txt file. However, you do NOT want to block anything that needs to be accessed in order to render the page so I would advise you to use the Robots.txt tester tool in GSC to make sure your Robots.txt code is not inadvertently blocking those image files as well, such as https://www.tidy-books.co.uk/skin/frontend/tidybooks/us/images/eco.png . As long as you aren't blocking access to the actual image files, only to the directory pages, your solution is OK.

It would be better to serve an error code on those pages though. Best practice is to not allow folder index pages to be loaded like that. Otherwise, it creates a bit of a security risk. Often webmasters will show a "403: Forbidden" code on those URLs.

I hope this helps. Please feel free to respond with any updates or clarifications. Hopefully others will chime in now that your issues are broken up into component parts. In the future, you'll get better responses by asking three different questions instead of lumping them all into one. This is because someone my know how to help with one of your questions, but not the others, and are less inclined to answer the one they can help with because they can't answer the others.

Good luck!

Hello Tom,

I think using RewriteRule is the way to go so no problem there. You may need to use a RewriteCond (condition) to exclude certain directories so you're not redirecting the entire site. You can read more here: http://stackoverflow.com/questions/4307780/whats-the-difference-between-rewriterule-and-redirect-301 .

I'll keep an eye on this and hopefully someone with more direct implementation experience with regex an help out.

As MrLeeB said, users may not be able to find what they're looking for and there would likely be some unnecessary wasting of "crawl budget", which could have an negative impact on indexation and rankings.

As Mike R. said, it would probably be terrible for the user-experience, and thus rankings and conversion rates, if you had too many categories.

For many years eCommerce businesses have built out category pages based on internal-searches performed by visitors, as well as the keywords they used to find the site on search engines. When used sparingly, this strategy can help inform the category structure of the site while offering a landing page for high-value keywords.

It can help copywriters understand which pain points and questions are trying to be solved, thus which should be called-out as benefits of the products within the category. It can even inform which products you source or develop to sell.

When abused, like when software, website code or database logic automates the process of creating these pages, the results can be devastating to the business.

After a few months of long-tail growth, the site may experience a major ranking loss across the entire domain for all but clearly branded and hyper-long-tail searches. If you are in the game of burn-n-churn, it might work for you provided you pay for some seed-traffic to build out the keyword lists.

...just in case you were thinking down that path.

No problem Cliff. I hope you found it helpful, and I'm happy to hear this has been cleared up for you. It's something I'll keep in mind if anyone else has a similar issue.

Hello CHolidays,

I don't see the pop-ups. Has this been changed since your post?

Often times on eCommerce sites you'll have a "quick view" window to see the item in a larger photo, and possibly read the product description. There is no single answer as to how this affects SEO because there are many ways to implement it. If I could see an example I would have more feedback.

I noticed while checking out the site that you have a broken link in the main navigation. It's the one that goes to the Island-Escapes page (Vacations).

Thanks Rhonda! It sounds like this thread will benefit many people then.

I'm sorry to hear that and can understand your frustration. Sometimes the way Google treats legitimate businesses in certain hyper-competitive industries is unfortunate. You're guilty until proven innocent.

I did not see anything wrong with your code. Maybe the manual reviewer didn't clear their cache? Maybe they were looking at a different URL than me, which still had the product schema without a buy button. Aggregate rating is clearly documented as being compatible with the Organization Schema type, and is described as "The overall rating, based on a collection of reviews or ratings, of the item."

If this were my site the next thing I would try is removing the Schema from every page except your home page. Maybe someone else has experienced this first-hand and will respond with a more helpful answer.

Ain't that the truth.

A good rule of thumb might be: If there's a BUY button that adds an item on that page to a shopping cart, use Product schema on that page. If that buy button is on every page of your site for the same product, it's going to seam suspicious coming from anyone but a household brand.

How do you know that wasn't it? I'm just curious so I can learn from it.

If you're using organization Schema, it should be OK to show on every page. If you're using Product schema it should only show on a page where they can buy the product.

I think it may have to do, as you guessed, with the use of Product schema. You should try Organization schema. If you have an actual product landing page, with a price and Add to Cart button, that would be where you would put the Product schema, but only put reviews about the actual reverse mortgage product. Most of the reviews on the home page seem to be about the company as a whole.

As long as you're using the same Schema in both places I don't think it would hurt to have them both - JSON-LD in the HTML header and as structured data markup in the HTML where that entity exists on the page.

You can play around with JSON-LD here to see what it would look like:
http://json-ld.org/playground/

Hello Sarmad,

I don't recommend having a single-page website on the front-end. You can handle the loading of content however you like: Javascript, lazy loading, dynamic serving.. as long as each section/page has a different URL. A hashtag is not a new URL, but rather a named anchor link to somewhere else on the same page. Thus, using this technique you would only have a single page indexed in Google, not a lot of room to target different topics.

Lots of sites use the parallax design style, but have multiple pages. Parallax and single-page are not necessarily synonymous. The most common use on a multi-page site would be to tell a story on a landing page or the home page.

More concerning to me right now is that it looks like they're putting their clients' sites on their own subdomains, which are fully indexable by Google: https://goo.gl/z9dSDl .

Webicultors,

Read this thread on Google's Product Forums. Let us know if it answers your question. If not, at least you're not alone...

Upon reading several similar threads on various forums and Q&A sites, it appears this is a very common occurrence resulting from a disparity between what the two tools define as an "error". The testing tool seems to be limited to errors in syntax / markup while GSC may see missing elements as errors.

Two servers? The existing one to process the redirects and a new one to handle the reverse proxy. Or vice versa. So the DNS for the old domain would point to a server that does the redirects. However, the server that hosts the site will be set to reverse proxy.

Another way of looking at this concept would be to take down and redirect the old site, and to start a new site, with the exact same files/database, that will be used to serve the content in the subdirectory/folder.

Ask what they think of that idea. I certainly don't have all the answers to every situations, but will do my best to help you find a workable solution.

Great answer Chris!

Manas,

It sounds to me like Google does not consider your brand to be an "entity" worth ranking for it's own brand name. This is why you're getting the "Did You mean?" link or "Search instead for..." in search results for your brand. The stronger your brand becomes - in Google's eye's - the less likely it is that people will see "Did You mean?" for the search.

Of course, without actually knowing the terms, it's difficult to say. If your brand name is "Helocopter" it would need to be VERY strong for Google not to show results for "Helicopter". However, if your brand is "HeelCooper" you could probably resolve that problem, and several of your others, with the suggestions below.

1. Go Through This Presentation and implement what you can, such as:

• Organization Schema with Schema.or Markup in the HTML or with JSON-LD

• Add and Define your brand on WikiData, Wikimedia and other open data sources, or repositories for brand entities

• Work your way up to WikiPedia by doing noteworthy things that generate press

3. Make sure your Name, Address and Phone Number (NAP) are consistent across the web.

• There are many ways to format things:

• (Street, St. | Road Rd. | 1800, 1-800 | 555-999-5555, (555) 999-5555) | ABC, ABZ.

• The important thing is consistency. You need to "Disambiguate" your brand from whatever that other keyword is. This is important for search, but also in reducing the amount of your potential customers who misspell your brand.

6. Drive more searches for your brand, and subsequent clicks to your site by generating positive publicity.

• Use PPC ads for your branded terms, and that other term if possible, to get as much of that traffic as possible to your site, even if you aren't ranking #1. Also, google will be less likely to recommend another search if the one you performed is generating income for them. And they can use the data gained from those real user searches to inform their algorithms, which will - hopefully - eventually result in your site showing up, as it should, for branded searches.

8. If none of this works, consider re-branding.

You need to do both.

The reverse proxy isn't about SEO so much as the ability to use subdirectories instead of subdomains, even though the subdirectories are hosted on different servers. You have to use a reverse proxy for the technological requirements of using a subdirectory (folder) instead of a subdomain when you're hosting sites on different servers but want to combine them on the same domain.

The 301 Redirects will ensure the users (including search engines) that visit the old URL on the subdomain get forwarded on to the new URL in the subdirectory. This is what preserves pagerank, and provides a good user experience. Do not keep the content available at the old URLs; Do not allow those URLs to return a 404 Status. 301 Redirect them.

I hope that clarifies your situation.

I checked the source with my default user agent (in this case Firefox) and did NOT see an H1 tag.

I checked with my user agent set to GoogleBot and DID see an H1 tag, which did have that keyword phrase in it.

I checked again with a default user agent, but this time with JavaScript disabled, and could not see anything at all on the viewable page (blank white page), though the source code was there without the H1 tag.

So it seems to me like you're pre-rendering the page for GoogleBot, and are including the H1 (and other header tags) as part of a fully-rendered page for search engines. However, because that Header tag does not exist if you turn JavaScript off - or if you're not Google - there may be a risk of Google seeing this page as "cloaking".

Pre-rendering is good. It's not a "bad" type of cloaking if you serve the EXACT same page to search engines that you serve to everyone else. Unfortunately, this does not seem to be the case with the way this page is set up. Google sees one thing, other visitors (with or without JavaScript enabled) see something else.

I know developers are head-over-heels for single-page apps and JavaScrpt frameworks, but this stuff is starting to drive me nuts. It's like trying to optimize Flash sites all over again. On the one hand you have Google bragging about how great they are at crawling JavaScript, even going so far as to say pre-rendering is not necessary... And on the other hand there are clear, sustained, organic search traffic drops whenever developers start turning flat HTML/CSS pages into these single-page JavaScript framework applications.

My advice to you is that if you're going to Pre Render a page for Google, to A: make sure the page a user with JavaScript enabled sees is exactly the same as what Google sees, and B: See if you can pre-render pages for visitors without JavaScript enabled as well.

Tormar,

Can you share the site so we can see exactly what's going on and provide a better answer?

We find many eCommerce sites overdo anchor text. For example, if you sell Car Stereos and you have brands/manufacturer categories in the navigation, this is too much anchor text for "car stereos" below:

Car Stereos by Brand
JVC Car Stereos
Clarion Car Stereos
Kenwood Car Stereos
Blaupunkt Car Stereos
....

We advice those clients to use labels like this:
Car Stereos by Brand
JVC
Clarion
Kenwood
Blaupunkt
...

Without seeing your site, I'm not sure exactly what you're doing. However, if you find that the changes you made continue to have a negative impact on your rankings, the obvious choice would be to undo those changes.

Hello 94501,

First of all, congratulations for being on top of this right now. It's unfortunate that most businesses are just trying to keep their heads above water on the basics. You are a very forward-thinking marketer.

a) Is this a good idea?
Yes, I do think this is a good idea. I've been experimenting with it myself.

b) What might help achieve the intended search result?
1. Domain Authority overall, as well as the Page Authority of these pages.
2. Being on the first page of Google, the higher the better (though they do use content from results lower on the page too)
3. Formatting on the page using easy-to-understand markup like header tags and lists, which it sounds like you're aware of already.
4. Formatting of the question and the answer to make that Q&A relationship obvious.

c) Does it matter that the box answers are summarized from a variety of folks forum posts, so not just one author?
1. I don't think the author, or mixture of authors, matters. However, it may be difficult to format it in such a way as to give multiple answers. Maybe choose the best one, or write your own based on an amalgamation of them all.

d) Might it drive up bounce rate on the actual page?
1. Not if it answers the question.
2. You risk someone not even coming to the page because they have their answer already, but I think this is what Google wants to happen so that wouldn't be a negative ranking signal. And if you're in spot #8 with a #1 Answer Box that's definitely a net gain for traffic.

e) Other considerations?
1. Have you looked into SEMRush yet? They have a feature in Beta right now that lets you see all of the enhanced search results (including Answer Box, Knowledge Graph and others) for which a site ranks.